Pew Research: Great Data, but the Analysis? P-ew!

The latest Pew Internet & American Life Project survey on search engine usage has been released (PDF).  (H/T: Joe Ferrara)  I think they provide an invaluable service in terms of gathering the data.  But their analysis ah… leaves much to be desired.  To wit:

WHY ARE MORE INTERNET USERS NOW INTEGRATING SEARCH INTO
THEIR ONLINE ACTIVITY?

While the number of internet users who search on a typical day has been steadily rising,
this is the second time since the Pew Internet Project began tracking search engine use
that we have seen a demonstrable leap in the numbers. The first came in late 2005, when
percentage of users searching on a typical day rose from about 30% (in June 2004) to
about 40% (in September 2005). We speculated at that time about a few possible reasons
behind the increase, pointing out that it was a time of much media coverage and buzz
about search engine companies, including the Google IPO.

Now, the percentage of users searching on a typical day has risen again, from about 40%
to 49%. What has changed in the search world that might account for this increase?

One likely reason is that users can now expect to find a high-performing, site-specific
search engine on just about every content-rich website that is worth its salt. With a
growing mass of web content from blogs, news sites, image and video archives, personal
websites, and more, internet users have an option to turn not only to the major search
engines, but also to search engines on individual sites, as vehicles to reach the
information they are looking for.

Another reason may be related to the fact that fully 55% of American homes have a highspeed
internet connection. Of all the demographic variables we analyzed, the presence of
a home broadband connection had the strongest relationship with a user’s propensity to
use a search engine on a typical day. Previous studies have shown that when a user
upgrades to home broadband, she is more likely to turn to the internet first when she has a question – and now she is increasingly likely to visit a search engine to find the answer.

Finally, it may be that general search engine sites have become so useful and well tuned
that people turn to them for an increasingly broad range of questions.

O rly?  As far as analysis goes, this is juvenile.  It’s like some high school book report.

First of all, “search engines on individual sites”?  When most people say “search engine”, they do not typically mean the little Search box at the top of most websites — including this here blog.  They mean Google, Yahoo!, and MSN, and the like.  If every act of searching for information is using a “search engine” then by golly, the Pew survey should also cover the Microsoft Windows Search tool that’s on every PC in the world.

Second, and more importantly, when Pew people say that a broadband user is more likely to turn to the Internet first… uh… what are the alternatives?  Encyclopedia Britannica?  Isn’t the real news not that 58% of home broadband users use search every day, but that 42% of them don’t?  I mean, what do these people do then instead of using search?

“No way, man — I’m telling you that Dan Marino has more career touchdowns than Joe Montana!”

“You’re crazy, bro, you don’t know what you’re talking about! Montana was the greatest.”

“Fine, let’s look it up.  I’ll start on this stack of Sports Illustrated back issues, and you start on that stack of Sporting News!”

“Oh man, if only there were a global network of computers that we could access using some sort of a digital connection, and some sort of an application that could take our query terms and match them up to database records… this would be much easier!”

As long as we’re being extraordinarily obvious, why not just say this as analysis of why more people are using search:

We think more people are using search, because they want to.

It’s about as useful in terms of insight.  The report itself is worth a read, even though it won’t surprise anyone who has not been living under the digital rock for the past five years.  But the analysis?  Pee-ewww.

-rsh